| dc.description.abstract | Intracranial haemorrhage has been reported as a cause of all kinds of strokes while
causing high mortality and disability for a longer duration and the accuracy of
volumetric measurement of the hematoma has a considerable impact on the correct
diagnosis and choice of treatments. This study aimed to compare a manual method
and a software-based method to measure hematoma volume while evaluating
changes in the mean HU value of the hematoma and the impact of the time interval
between the two consecutive scans on hematoma volume using non-contrast
computed tomography (NCCT) head images. The volumetric measurement of each
hematoma was carried out separately using a manual method and a software-based
(SB) method in 61 NCCT head images. For the manual method, TADA formula was
used to estimate the hematoma volume with length, width and depth measured on
NCCT axial and sagittal brain images. 3D slicer (Slicer 4.10.2) software was used for
the software-based volume analysis and to measure the HU value of the hematoma.
Two volume measurement methods demonstrated a significant difference between
their measurements (p<0.05) with a significant strong positive correlation with
each(r=0.977). Furthermore, hematoma volume difference and time interval
between scans revealed a moderate positive correlation(r=0.517) and the
haemoglobin level at admission demonstrated mild positive correlations with
hematoma volumes (r=0.274) both of which are statistically significant. Significant
mild positive correlations were found between mean HU value and the two
hematoma volumes (r= 0.307, r= 0.263) respectively, whilst the mean HU difference
demonstrated a moderate positive correlation with hematoma volume difference.
The TADA approach demonstrated a significant difference in volume measurements
when compared to a software-based method, therefore more research is needed to
support the implementation of an internally designed automatic volumetric
measuring method to improve accuracy and efficacy. | en_US |
